Tomographic Defect Detection via Principal Component Subtraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting metallurgical anomalies in high-pressure turbine blades using tomographic images face challenges due to local variations in gray levels and the complexity of distinguishing defects from the background, leading to difficulties in automatic detection and high production costs for non-destructive testing.
Innovation Solution
A method involving the creation of a reference base from healthy parts using decomposition into principal components, allowing for the segmentation and identification of defects by subtracting a synthetic image representing a healthy part from the actual part image, thereby highlighting anomalies and reducing data variability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If tomographic imaging is used to reconstruct the blade in three dimensions, then the ability to detect internal defects is improved, but the complexity of data analysis increases significantly due to the large volume of information
Solution Approach 1:
The patent applies segmentation by dividing the three-dimensional tomographic image into multiple two-dimensional slices or sections. This allows the large volume of data to be processed in smaller, manageable units while preserving the ability to detect defects in three-dimensional space. The segmentation reduces computational complexity by breaking down the analysis into sequential two-dimensional operations rather than attempting to process the entire three-dimensional volume simultaneously.
2Extent of automation
If standard detection methods are used on tomographic images, then the process can be automated, but local variations in gray levels and background complexity make defect detection inaccurate
Solution Approach 1:
The patent applies local quality by analyzing each local region of the tomographic image according to its specific characteristics rather than applying a uniform detection threshold throughout. The method compares gray level variations in each local area against expected patterns for that specific region, accounting for local background variations. This allows automatic detection to maintain high accuracy by adapting to local conditions while still achieving full automation.
3Strength
If more complex circuits are designed in turbine blades to optimize thermomechanical strength, then the mechanical performance is improved, but the radiographic inspection becomes more complicated due to superpositions of patterns
Solution Approach 1:
The patent applies dimensionality change by transitioning from two-dimensional radiographic inspection to three-dimensional tomographic imaging. This additional dimension allows the inspection system to distinguish between features at different depths, resolving the superposition problems that occur in conventional radiography. The complex circuits embedded in the blade can be visualized and inspected in three-dimensional space, separating overlapping patterns that would be indistinguishable in two-dimensional images.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables effective detection of defects in three-dimensional parts by reducing data complexity and enhancing the visibility of anomalies, thereby improving inspectability and reducing production costs through automated inspection processes.
Implementation Method 1
X-ray inspection systems are central to the quality control since they allow a detection of internal defects
Implementation Method 2
the value of the points of which depends on the density and on the absorption factor of the material
Implementation Method 3
imaging by tomography. This method allows to reconstruct the blade in three dimensions in a tensor
Data Source
AI summary
A method for determining the presence of at least one defect in a three-dimensional part on the basis of a reference base, wherein:an image of the part is acquired by tomography;the image is segmented and the segmented image is registered with respect to the images of the reference base;the subdomains associated with the registered segmented image are determined; andfor each subdomain of the registered segmented image, the following steps are carried out:registering the subdomain;breaking down the subdomain with respect to the principal modes of the reference base;generating a synthetic image of the subdomain from the breakdown, which is subtracted from the registered segmented image of the subdomain; anddetermining that a defect is present on the basis of the comparison of the value of each pixel of the residual image with a predetermined threshold.